Life and Time - PxVx - Bivariate Models

Intro

Template for model comparisons:

The above diagram captures all 4 possible models combinations. The black is the univariate ALT model, and the blue adds all paths we estimate for the bivariate ALT. Red paths show the paths constrained to be 0 in the mean-only (fixed slope) model. Letter labels indicate paths constrained to be equal.

What are the best models that converge?

In the below figures, each bar represents the ‘model score’ of a model combination for a bivariate ALT model. A score of 2 indicates that there were no problems, 1 indicates that there were warnings about the standard error being computed (e.g., with MLF instead of MLR), and 0 indicates that there were errors (non-convergence).

The tables show the winning model giving preference to the full bivariate random slope ALT model, even if it was estimated with MLF instead of MLR (this is not very common in the national sample, but more so in the college sample).

The model for the personality variable comes first, so ‘Lin_MeanOnly’ means ‘full linear ALT model for personality variable, fixed mean slope ALT model for value variable.’

National

pVar	aspfin	BFA_MT	HRZ_COL	HRZ_IND	MVS_mc	USI	VRT_COL	VRT_IND
BFA_AC	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_AP	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_CI	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_CO	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_EA	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_EE	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_NV	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_NW	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_OI	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_OO	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_A6	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_C	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_E	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
bfi_hp8	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_N	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_O	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
D_SCALE	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
I_A	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
I_C	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly	Lin_Lin	Lin_Lin	Lin_Lin
I_D	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
I_E	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly	Lin_Lin	Lin_Lin	Lin_Lin
I_N	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly	Lin_Lin	Lin_Lin	Lin_Lin
I_O	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly	Lin_Lin	Lin_Lin	Lin_Lin
I_S	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
S_SCALE	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin

National Sample

College

pVar	aspfin	BFA_MT	HRZ_COL	HRZ_IND	MVS_mc	USI	VRT_COL	VRT_IND
BFA_AC	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_AP	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_CI	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	MeanOnly_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_CO	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin
BFA_EA	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_EE	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_NV	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_NW	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_OI	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFA_OO	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_A6	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_C	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly	Lin_Lin	Lin_Lin	Lin_Lin
BFI_E	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly	Lin_Lin	Lin_Lin	Lin_Lin
bfi_hp8	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
BFI_N	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin
BFI_O	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
D_SCALE	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin
S_SCALE	Lin_Lin	Lin_Lin	Lin_Lin	Lin_Lin	Lin_MeanOnly MeanOnly_Lin	Lin_Lin	Lin_Lin	Lin_Lin

College Sample

Non-Lin_Lin Models: Univariate best-fit

sample	pVar	vVar	bestCombo	AIC_P	BIC_P	LL_P	AIC_V	BIC_V	LL_V
Col	BFA_CI	MVS_mc	MeanOnly_Lin	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFA_CO	VRT_IND	Lin_MeanOnly MeanOnly_Lin	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFA_EE	MVS_mc	Lin_MeanOnly MeanOnly_Lin	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFA_NW	MVS_mc	Lin_MeanOnly MeanOnly_Lin	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFA_OI	MVS_mc	Lin_MeanOnly MeanOnly_Lin	AR_LinM	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFI_A6	MVS_mc	Lin_MeanOnly MeanOnly_Lin	AR_LinM	AR_LinM	AR_LinM	AR_Int	AR_Int	AR_Int
Col	BFI_C	MVS_mc	Lin_MeanOnly	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFI_E	MVS_mc	Lin_MeanOnly	AR_LinM	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	BFI_N	VRT_IND	Lin_MeanOnly MeanOnly_Lin	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Col	S_SCALE	MVS_mc	Lin_MeanOnly MeanOnly_Lin	AR_Lin	AR_Int	AR_Int	AR_Int	AR_Int	AR_Int
Nat	I_C	MVS_mc	Lin_MeanOnly	AR_LinM	AR_LinM	AR_LinM	AR_Lin	AR_Lin	AR_Lin
Nat	I_E	MVS_mc	Lin_MeanOnly	AR_LinM	AR_Int	AR_LinM	AR_Lin	AR_Lin	AR_Lin
Nat	I_N	MVS_mc	Lin_MeanOnly	AR_LinM	AR_Int	AR_LinM	AR_Lin	AR_Lin	AR_Lin
Nat	I_O	MVS_mc	Lin_MeanOnly	AR_Int	AR_Int	AR_Int	AR_Lin	AR_Lin	AR_Lin

Parameter plots

National: AR paths across variable combinations and model types

National: XL paths across variable combinations and model types

College: AR paths across variable combinations and model types

College: XL paths across variable combinations and model types

National: Within-wave correlations across variable combinations and model types

College: Within-wave correlations across variable combinations and model types